INDEX

Explanations

names starting with "Ma" and potentially related words or phrases

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

side

-0.69

suit

-0.68

tap

-0.68

sed

-0.65

lined

-0.63

deck

-0.62

 hazard

-0.62

dem

-0.62

AAP

-0.62

ĳå£«

-0.62

POSITIVE LOGITS

estro

1.40

ureen

1.33

ples

1.15

arten

1.02

Ã±

1.01

isel

0.99

plin

0.96

illard

0.95

pling

0.93

iami

0.92

Activations Density 0.023%

names starting with "Ma" and potentially related words or phrases

No Comments

No Known Activations